Effects of pressure, contact time, permeance, and selectivity in membrane reactors: The case of the dehydrogenation of ethane

This paper provides some general conclusions about the effects of pressure, contact time, permeance, and selectivity in membrane reactors for the dehydrogenation of ethane. It is demonstrated that elevated pressure is desirable for the operation of membrane reactors even though the equilibrium may be unfavorable because of a net increase in moles (such as occurs in many hydrogen-producing reactions). This is because the increased reactant throughputs at higher pressures more than compensates for the decreased conversion, and because permeance is favored with higher pressure differences. It is also shown that a permeance above 10(-7) mol m(-2) s(-1) Pa-1 is needed for satisfactory operation and that product selectivities above 100 do not substantially improve performance. Finally, it is shown that contact times above 10 s are important for operating membrane reactors. As a model A -> B + C reaction, the dehydrogenation of ethane (C2H6 -> C2H4 + H-2) was studied over a 5 wt % Cr/ZSM-5 catalyst in a conventional packed-bed reactor (PBR) and in a membrane reactor (MR) fitted with hydrogen selective silica membranes at various feed flow rates, total pressures and temperatures. Two types of membranes were used. One was prepared by the chemical vapor deposition (CVD) of vinyltriethoxysilane (VTES) at 873K and had a high H-2 permeance of 4.0 x 10(-7) mol m(-2) S-1 Pa-1 with a H-2/C2H6 selectivity of 240 at 823 K. The other was prepared by the CVD of a mixture of silica and zirconia precursors, and had a lower H-2 permeance of 1.9 x 10(-7) mol m(-2) s(-1) Pa-1 but a higher H-2/C2H6 selectivity of 4300 at 823 K. At all reaction conditions the conversion of C2H6 and yields of C2H4 and H-2 in the MR were higher than those in the PBR. The effect of pressure was studied and an important result was the finding that the productivity in H-2 formation increased with increasing pressure even though the conversion of C2H6 decreased for thermodynamic reasons. A one-dimensional model was used to describe the MR, and it's performance was evaluated by correlating the yield enhancement of C2H4 with the permeance of the membranes and also a parameter called an operability level coefficient (OLC), which is the ratio between the permeation rate and the formation rate of a H-2 in the membrane reactor. Another notable finding was that a H-2 permeance selectivity of 100 was sufficient to achieve good performance in the MR. Higher selectivities did not improve H-2 yields significantly.